What is the status on multi-core CPU's working on a single task?

[DanDaMan315]

Like to know what the update is.

 

[MarcVenice]

Define single task? Single application? If the app is multithreaded it will use both cores, if it isn't, it won't ...

 

[nomagic]

Maybe he is talking about something else? A lot of work has been done to make software multi-core friendly.

programming model, hardware, compiler....

His question is too broad. In fact, I think it is a little insulting. People should not ask question this general and expect to be answered properly.

 

[Denithor]

I hope you aren't asking about those two-year-old rumors of "reverse hyperthreading" that would 'combine' two processors into one for single-threaded applications.

 

[Modelworks]

If you are talking about a cpu that can take a single calculation and divide it between all available cores without the software being involved it is still in its infancy.
The embedded market has some devices that can do it but I don't expect we will see it in the pc market anytime soon. There is one processor by parallax that is in development that takes all task and splits it up between 8 cores.

This is the current design.
http://www.parallax.com/Portal...kDiagram-RevAhoriz.pdf

The next will do the division of task automatically without the software having to do it.

 

[soccerballtux]

Originally posted by: nomagic
Maybe he is talking about something else? A lot of work has been done to make software multi-core friendly.

programming model, hardware, compiler....

His question is too broad. In fact, I think it is a little insulting. People should not ask question this general and expect to be answered properly.

Reminds me of a guy at work who would look at you and say "I need your help. I need to test the network." And then expect you to magically understand everything that needs to be done and tell him the list of steps for how to do this.

 

[Idontcare]

Originally posted by: DanDaMan315
Topic Title: What is the status on multi-core CPU's working on a single task?

Like to know what the update is.

The status is that they 11.6% complete with the majordomo project, feasibility timeline has reached nucleation saturatibility and are accelerating towards an end-completion milestone with better than parity throughput.

Will debrief another update post haste.

 

[waffleironhead]

Originally posted by: Idontcare

Originally posted by: DanDaMan315
Topic Title: What is the status on multi-core CPU's working on a single task?

Like to know what the update is.

The status is that they 11.6% complete with the majordomo project, feasibility timeline has reached nucleation saturatibility and are accelerating towards an end-completion milestone with better than parity throughput.

Will debrief another update post haste.

How did they manage to calibrate the dangleshanks? I always figured this was the major blockage point on passing that crucial 8.5% milestone.

 

[sonoran]

Originally posted by: DanDaMan315
Like to know what the update is.

Like to know why you're still running MS-DOS.

 

[Cogman]

Originally posted by: waffleironhead

Originally posted by: Idontcare

Originally posted by: DanDaMan315
Topic Title: What is the status on multi-core CPU's working on a single task?

Like to know what the update is.

The status is that they 11.6% complete with the majordomo project, feasibility timeline has reached nucleation saturatibility and are accelerating towards an end-completion milestone with better than parity throughput.

Will debrief another update post haste.

How did they manage to calibrate the dangleshanks? I always figured this was the major blockage point on passing that crucial 8.5% milestone.

Well, physicists found that by using reverse quantum tunneling, dangleshanks could easily be circumvented. Now the biggest problem is that the capacitor in the flux section of the cpu is overheating with too much Mesinbalm radiation. Right now they are testing F32 to see if it will circumvent this problem. Expect results with in the next solstice. I expect the 14.2% milestone to be announced then.

 

[Idontcare]

Originally posted by: Cogman

Originally posted by: waffleironhead

Originally posted by: Idontcare

Originally posted by: DanDaMan315
Topic Title: What is the status on multi-core CPU's working on a single task?

Like to know what the update is.

The status is that they 11.6% complete with the majordomo project, feasibility timeline has reached nucleation saturatibility and are accelerating towards an end-completion milestone with better than parity throughput.

Will debrief another update post haste.

How did they manage to calibrate the dangleshanks? I always figured this was the major blockage point on passing that crucial 8.5% milestone.

Well, physicists found that by using reverse quantum tunneling, dangleshanks could easily be circumvented. Now the biggest problem is that the capacitor in the flux section of the cpu is overheating with too much Mesinbalm radiation. Right now they are testing F32 to see if it will circumvent this problem. Expect results with in the next solstice. I expect the 14.2% milestone to be announced then.

I wasn't aware that we were allowed to publicly discuss the 14.2% milestone. I musta missed that memo, prolly busy busting out that TPS coversheet on my weekly. Goodz 2 knowz.

 

[Janooo]

Originally posted by: Idontcare

Originally posted by: Cogman

Originally posted by: waffleironhead

Originally posted by: Idontcare

Originally posted by: DanDaMan315
Topic Title: What is the status on multi-core CPU's working on a single task?

Like to know what the update is.

The status is that they 11.6% complete with the majordomo project, feasibility timeline has reached nucleation saturatibility and are accelerating towards an end-completion milestone with better than parity throughput.

Will debrief another update post haste.

How did they manage to calibrate the dangleshanks? I always figured this was the major blockage point on passing that crucial 8.5% milestone.

Well, physicists found that by using reverse quantum tunneling, dangleshanks could easily be circumvented. Now the biggest problem is that the capacitor in the flux section of the cpu is overheating with too much Mesinbalm radiation. Right now they are testing F32 to see if it will circumvent this problem. Expect results with in the next solstice. I expect the 14.2% milestone to be announced then.

I wasn't aware that we were allowed to publicly discuss the 14.2% milestone. I musta missed that memo, prolly busy busting out that TPS coversheet on my weekly. Goodz 2 knowz.

Mutex issue on your laptop serial port caused the exchange server to mix up email addresses of the above mentioned memo. Time to throw your dial up out of the windows or switch to Linux.

 

[aigomorla]

are u refering to the old system that AMD propsed in which a dualcore could in essense do the same work on a single task and you could double the core speed because of it?

Because i think that was can'd a very long time ago, and you'll never see it, because first off if a program does require just only 1 task cpu, then its not a very powerful program which would need to be run even faster with today's processors.

For instance i have a program that will use all 16 threads my new system will have.

In short, programers found it easier to program for multi core then hardware designers figuring how to merg 2 cores together.